This project will improve the performance and utility of centrifugal ultra filtration now widely used for laboratory concentration and desalting of macromolecules. A 7 mL prototype device having regenerated cellulose (RC) non-fouling membrane area 4x greater than current 2mL RC devices has been shown to have l00x greater rate in concentrating 0.25% cytochrome c (patents pending). Furthermore, 12 ug cytochrome c in 5 mL was concentrated 30 min to a mean volume of 5. 5uL Mean direct pipette protein recovery was 79%. A 1mL wash recovered a mean of 11% more. This is also better than the 50% reported for conventional RC devices at this concentration. The first goal is to scale down this invention one order of magnitude to efficiently process sample volumes < 1mL. The second goal is to exploit the larger area for fractionation by optimizing pressure (unit flux) and exchange volume to balance process time vs increased selectivity. Model systems will include removal of oligonucleotides and dye terminators from DNA, removal of oligomers from cytochrome c, and removal of albumin and larger globulins from Prostate Specific Antigen in plasma. The third goal is to develop a multiwell strip version of the invention for processing multiples of 12 samples. PROPOSED COMMERCIAL APPLICATION: Laboratory centrifugal concentrators which are faster and give higher recovery will provide immediate value to established applications in protein chemistry and moleclular biology. Improved speed will be of most value for use with smaller molecules where times of more than one hour are needed with current devices. Ability to operate more efficiently and in parallel arrays will benefit emerging applications in proteomics. Fractionation by ultra filtration would open new research applications and expand this mature market.